[GH-ISSUE #6] 能否集成到vscode的插件里 #6

Open
opened 2026-10-26 21:01:57 +00:00 by bsakaguchi · 10 comments
Owner

Originally created by @Lyan-ing on GitHub (Nov 3, 2023).
Original GitHub issue: https://github.com/deepseek-ai/DeepSeek-Coder/issues/6

如题^_^

Originally created by @Lyan-ing on GitHub (Nov 3, 2023). Original GitHub issue: https://github.com/deepseek-ai/DeepSeek-Coder/issues/6 如题^_^
Author
Owner

@yportne13 commented on GitHub (Nov 3, 2023):

可以魔改 codeshell 的 vscode 插件 ^^ 我试过用 llama.cpp 的 server 起一个 mistral-7b 搭配 codeshell vscode 插件,只需要修改插件内置的 prompt 改成 mistral 的格式就能正常使用了 ^^

<!-- gh-comment-id:1791813253 --> @yportne13 commented on GitHub (Nov 3, 2023): 可以魔改 codeshell 的 vscode 插件 ^_^ 我试过用 llama.cpp 的 server 起一个 mistral-7b 搭配 codeshell vscode 插件,只需要修改插件内置的 prompt 改成 mistral 的格式就能正常使用了 ^_^
Author
Owner

@opptimus commented on GitHub (Nov 3, 2023):

同问

<!-- gh-comment-id:1791945845 --> @opptimus commented on GitHub (Nov 3, 2023): 同问
Author
Owner

@guoday commented on GitHub (Nov 4, 2023):

目前正在集成中,但是可以使用开源的插件如refact,替换成我们的模型即可

<!-- gh-comment-id:1793313337 --> @guoday commented on GitHub (Nov 4, 2023): 目前正在集成中,但是可以使用开源的插件如refact,替换成我们的模型即可
Author
Owner

@x4080 commented on GitHub (Nov 9, 2023):

@yportne13 Where can I change the built in prompt from codeshell vscode ? Thanks

Edit : When using regular llama cpp with deepseek coder, when completion there's message

type error: respData.split is not a function
<!-- gh-comment-id:1804722221 --> @x4080 commented on GitHub (Nov 9, 2023): @yportne13 Where can I change the built in prompt from codeshell vscode ? Thanks Edit : When using regular llama cpp with deepseek coder, when completion there's message ``` type error: respData.split is not a function ```
Author
Owner

@hobpond commented on GitHub (Nov 28, 2023):

Have you tried Continue.dev, an extension for VSCode? You can load the model with llama.cpp and add Deepseek as the model to use.
Here's my config (in the continue.dev text box, type /config to access) Got it from YearZero on TheBloke's discord.

{
  "models": [
    {
      "title": "CodeLlama-34b-Instruct",
      "provider": "llama.cpp",
      "model": "codellama-34b",
      "api_base": "http://localhost:8080"
    },
    {
      "title": "CodeLlama-7b-Instruct",
      "provider": "llama.cpp",
      "model": "codellama-7b",
      "api_base": "http://localhost:8080",
      "system_message": ""
    },
    {
      "title": "DeepSeek",
      "provider": "llama.cpp",
      "model": "deepseek-33b",
      "api_base": "http://localhost:8080"
    }
  ],
  "model_roles": {
    "default": "DeepSeek",
    "chat": "DeepSeek",
    "edit": "DeepSeek",
    "summarize": "DeepSeek"
  },
  "system_message": "",
  "slash_commands": [
    {
      "name": "edit",
      "description": "Edit highlighted code",
      "step": "EditHighlightedCodeStep"
    },
    {
      "name": "config",
      "description": "Customize Continue",
      "step": "OpenConfigStep"
    },
    {
      "name": "comment",
      "description": "Write comments for the highlighted code",
      "step": "CommentCodeStep"
    },
    {
      "name": "clear",
      "description": "Clear step history",
      "step": "ClearHistoryStep"
    },
    {
      "name": "share",
      "description": "Download and share this session",
      "step": "ShareSessionStep"
    },
    {
      "name": "cmd",
      "description": "Generate a shell command",
      "step": "GenerateShellCommandStep"
    }
  ],
  "custom_commands": [
    {
      "name": "test",
      "prompt": "Write a comprehensive set of unit tests for the selected code. It should setup, run tests that check for correctness including important edge cases, and teardown. Ensure that the tests are complete and sophisticated. Give the tests just as chat output, don't edit any file.",
      "description": "Write unit tests for highlighted code"
    }
  ],
  "context_providers": [
    {
      "name": "diff"
    },
    {
      "name": "url"
    },
    {
      "name": "terminal"
    }
  ]
}
<!-- gh-comment-id:1830402031 --> @hobpond commented on GitHub (Nov 28, 2023): Have you tried Continue.dev, an extension for VSCode? You can load the model with llama.cpp and add Deepseek as the model to use. Here's my config (in the continue.dev text box, type /config to access) Got it from YearZero on TheBloke's discord. ```JSON { "models": [ { "title": "CodeLlama-34b-Instruct", "provider": "llama.cpp", "model": "codellama-34b", "api_base": "http://localhost:8080" }, { "title": "CodeLlama-7b-Instruct", "provider": "llama.cpp", "model": "codellama-7b", "api_base": "http://localhost:8080", "system_message": "" }, { "title": "DeepSeek", "provider": "llama.cpp", "model": "deepseek-33b", "api_base": "http://localhost:8080" } ], "model_roles": { "default": "DeepSeek", "chat": "DeepSeek", "edit": "DeepSeek", "summarize": "DeepSeek" }, "system_message": "", "slash_commands": [ { "name": "edit", "description": "Edit highlighted code", "step": "EditHighlightedCodeStep" }, { "name": "config", "description": "Customize Continue", "step": "OpenConfigStep" }, { "name": "comment", "description": "Write comments for the highlighted code", "step": "CommentCodeStep" }, { "name": "clear", "description": "Clear step history", "step": "ClearHistoryStep" }, { "name": "share", "description": "Download and share this session", "step": "ShareSessionStep" }, { "name": "cmd", "description": "Generate a shell command", "step": "GenerateShellCommandStep" } ], "custom_commands": [ { "name": "test", "prompt": "Write a comprehensive set of unit tests for the selected code. It should setup, run tests that check for correctness including important edge cases, and teardown. Ensure that the tests are complete and sophisticated. Give the tests just as chat output, don't edit any file.", "description": "Write unit tests for highlighted code" } ], "context_providers": [ { "name": "diff" }, { "name": "url" }, { "name": "terminal" } ] } ```
Author
Owner

@x4080 commented on GitHub (Nov 28, 2023):

@hobpond Thanks, I'll try it

<!-- gh-comment-id:1830774323 --> @x4080 commented on GitHub (Nov 28, 2023): @hobpond Thanks, I'll try it
Author
Owner

@wwwzhouhui commented on GitHub (Dec 14, 2023):

I have a plan to implement it using codegpt+DeepSeek-Coder
https://www.youtube.com/watch?v=cAfWOMZXPGk&t=338s
https://www.bilibili.com/video/BV1EC4y1X74C/

<!-- gh-comment-id:1855084185 --> @wwwzhouhui commented on GitHub (Dec 14, 2023): I have a plan to implement it using codegpt+DeepSeek-Coder https://www.youtube.com/watch?v=cAfWOMZXPGk&t=338s https://www.bilibili.com/video/BV1EC4y1X74C/
Author
Owner

@x4080 commented on GitHub (Dec 14, 2023):

@wwwzhouhui cool !

<!-- gh-comment-id:1856556317 --> @x4080 commented on GitHub (Dec 14, 2023): @wwwzhouhui cool !
Author
Owner

@FunkyGod commented on GitHub (Aug 27, 2024):

有什么好的简单方案继承deepseek到vscode吗?

<!-- gh-comment-id:2311712621 --> @FunkyGod commented on GitHub (Aug 27, 2024): 有什么好的简单方案继承deepseek到vscode吗?
Author
Owner

@itdir commented on GitHub (Mar 3, 2025):

可以魔改 codeshell 的 vscode 插件 ^^ 我试过用 llama.cpp 的 server 起一个 mistral-7b 搭配 codeshell vscode 插件,只需要修改插件内置的 prompt 改成 mistral 的格式就能正常使用了 ^^

Ni hao ma?

Something might've been lost in the translation.

Xie xie!

<!-- gh-comment-id:2695512949 --> @itdir commented on GitHub (Mar 3, 2025): > 可以魔改 codeshell 的 vscode 插件 ^^ 我试过用 llama.cpp 的 server 起一个 mistral-7b 搭配 codeshell vscode 插件,只需要修改插件内置的 prompt 改成 mistral 的格式就能正常使用了 ^^ Ni hao ma? Something might've been lost in the translation. Xie xie!
Sign in to join this conversation.
No milestone
No project
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
deepseek-ai/DeepSeek-Coder#6
No description provided.